This function should be invisible to most users, and is part of our the fitting routine using the EM algorithm. Our maximum likelihood procedure uses an iterative algorithm called Expectation-Maximization. This requires a starting point, chosen at random. EM.starting point randomly assigns this starting point.

1 | ```
EM.starting.point(d, trait = "binary")
``` |

`d` |
The dataframe that needs to be initialized |

`trait` |
Can be either “binary” or ”eQTL” |

Returns the input data frame with reasonable random starting values.

Vincent Plagnol vincent.plagnol@cimr.cam.ac.uk and Chris Barnes christopher.barnes@imperial.ac.uk

A. P. Dempster, N. M. Laird, and D. B. Rubin, "Maximum likelihood from incomplete data via the EM algorithm," J. Royal Stat. Soc., vol. 39, pp. 1–38, 1977.

Questions? Problems? Suggestions? Tweet to @rdrrHQ or email at ian@mutexlabs.com.

Please suggest features or report bugs with the GitHub issue tracker.

All documentation is copyright its authors; we didn't write any of that.